Abstract P142: The Neurologic Examination of Cardiac Arrest Patients During the Post-hypothermia Era is Insufficient to Exclude Survival
Background: Prognosticating neurological recovery after CPR uses decision rules based on relatively old data. Validity of these rules in the modern ICU, especially with implementation of hypothermia is unknown. This study estimated the accuracy of traditional prognostic criteria of pupil, corneal, and motor response under current practice.
Methods: Survival and neurologic outcome were determined from hospital records for all cardiac arrest patients treated in a single tertiary care hospital between 1–1– 05 and 3–30 – 09. Motor component of the Glasgow Coma Scale (GCS-M) and eye examinations were recorded at 24 and 72 hours after arrest. Exams in presence of sedation were considered ‘missing.’ Odds ratios (OR and 95% confidence intervals, CI) for survival were calculated based on motor scores, pupil, and corneal responses. Survival rates (and 95% CI) were calculated for subjects with traditionally “poor” exams (GCS-M<<26>flexion, 3; absent pupil or corneal response).
Results: From 406 subjects hospitalized after CPR, 256 were comatose, 158 (62%) received hypothermia, and 45 died prior to 72 hours. Mean age (61, SD16 years), gender (147 male; 57%), out-of-hospital collapse (163, 64%), and primary ventricular dysrhythmia (92, 36%) in the comatose group resembled prior series. Survival was 82/256 (32%) with 46/256 (18%) good outcomes. Valid motor data were available for 197 and 138 cases at 24 and 72 hours. Among cases with GCS-M <<26>3 at 24 hours or at 72 hours, 19/92 (21%) and 8/49 (16%) lived including 5 with no motor response at 72 hours. OR for survival with GCS-M<<26>3 at 24 and 72 hours were 0.22 (95%CI .13, .38) and 0.10 (95%CI .03, .33). Valid eye data were available for 143 and 85 cases at arrival and 72 hours. Among cases with no pupil reaction at arrival and 72 hours, 6/50 (12%) and 22/36 (61%) lived. Among cases with no corneal response at 72 hours, 0/22 (0%) lived. OR for survival with pupil response at arrival and 72 hours were 14.7 (95%CI 5.64, 38.1), 30.5 (95%CI 3.8, 247.1).
Conclusions: In modern practice, overall survival is higher than in benchmark series used for prognostication. While GCS-M<<26>3 (flexor posturing), absent pupilary responses and corneal responses are all associated with lower odds of survival, these signs do not exclude survival in the modern ICU.